Bias in the Tails: How Name-conditioned Evaluative Framing in Resume Summaries Destabilizes LLM-based Hiring
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915948484624384 |
|---|---|
| author | Nghiem, Huy Nguyen-Le, Phuong-Anh Ho, Sy-Tuyen Daume III, Hal |
| author_facet | Nghiem, Huy Nguyen-Le, Phuong-Anh Ho, Sy-Tuyen Daume III, Hal |
| contents | Research has documented LLMs' name-based bias in hiring and salary recommendations. In this paper, we instead consider a setting where LLMs generate candidate summaries for downstream assessment. In a large-scale controlled study, we analyze nearly one million resume summaries produced by 4 models under systematic race-gender name perturbations, using synthetic resumes and real-world job postings. By decomposing each summary into resume-grounded factual content and evaluative framing, we find that factual content remains largely stable, while evaluative language exhibits subtle name-conditioned variation concentrated in the extremes of the distribution, especially in open-source models. Our hiring simulation demonstrates how evaluative summary transforms directional harm into symmetric instability that might evade conventional fairness audit, highlighting a potential pathway for LLM-to-LLM automation bias. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_19984 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Bias in the Tails: How Name-conditioned Evaluative Framing in Resume Summaries Destabilizes LLM-based Hiring Nghiem, Huy Nguyen-Le, Phuong-Anh Ho, Sy-Tuyen Daume III, Hal Computers and Society Artificial Intelligence Computation and Language Research has documented LLMs' name-based bias in hiring and salary recommendations. In this paper, we instead consider a setting where LLMs generate candidate summaries for downstream assessment. In a large-scale controlled study, we analyze nearly one million resume summaries produced by 4 models under systematic race-gender name perturbations, using synthetic resumes and real-world job postings. By decomposing each summary into resume-grounded factual content and evaluative framing, we find that factual content remains largely stable, while evaluative language exhibits subtle name-conditioned variation concentrated in the extremes of the distribution, especially in open-source models. Our hiring simulation demonstrates how evaluative summary transforms directional harm into symmetric instability that might evade conventional fairness audit, highlighting a potential pathway for LLM-to-LLM automation bias. |
| title | Bias in the Tails: How Name-conditioned Evaluative Framing in Resume Summaries Destabilizes LLM-based Hiring |
| topic | Computers and Society Artificial Intelligence Computation and Language |
| url | https://arxiv.org/abs/2604.19984 |